Stackoverflow and Stack Exchange are usually hosted in that building at peer 1, but we failed over to our secondary dc out in OR ealier in the evening.

It's interesting to see how a storm can singlehandedly cut out some major websites and cause millions to go without power. I think it shows how fragile this tech world really is, and how little we should rely on it in case of a major disaster.

Here's the part I don't understand. This was a record 13 foot storm surge. Is it true that, had the pumps been located above the storm surge, service would have continued uninterrupted? It sounds like everything else in their data center is just fine.

This makes it seem like there was a serious design flaw. These pumps were a critical part of the system, but were located in a vulnerable area.

F*$cked

Ironic they tweeted this re AWS 12h ago: Internap ‏@Internap Could 'Frankenstorm' Lead To Another AWS #Outage? http://onforb.es/Vtr45J

I really don't know what the proper procedure should be in a data center when the big klaxon goes off and you here those fateful words over the intercom, "Close all doors, DIVE, DIVE, DIVE!"

Clouds are bad for cloud computing.

Update at 10:17ET http://pastebin.com/NUQNHHJi Interesting they don't have a public status page of any sort.

Lifehacker/Gizmodo and its affiliate sites are down too

Are they unable to bypass the destroyed pumps and take fuel to the generators via 55-gallon drums? If so, why not?

See also https://twitter.com/DEVOPS_PORN

So they have more sense than DirectNIC had during Katrina http://www.wired.com/science/discoveries/news/2005/09/68725

Internap LGA11 lost power at 11:48AM ET http://pastebin.com/6AxvbzF1

Some websites affected by catastrophe: OccupyWallSt.org, Alternet.org

Internap's cloud service taken down by Sandy's cloud service.

undefined